Borderline personality disorder is a serious mental illness characterized by extreme emotions, chaotic interpersonal relationships, suicidal behaviors, and a poor sense of self. Offspring of mothers with Borderline Personality Disorder (BPD) are at an elevated risk for developing mental illness across the lifespan. Difficulty managing emotions are a hallmark feature of BPD, and yet the ability to do so is necessary for responding effectively to childrens' emotions. This process is called maternal emotion socialization, which has a major impact on how children develop their own emotion regulation (ER) skills. ER develops rapidly during preschool and deficits in preschool ER are recognized as underlying future mental disorders. This proposal will test a model examining the extent to which maternal ER and emotion socialization impact child ER, which may be a significant pathway by which mental health problems are transmitted to offspring of mothers with BPD. This proposal will leverage Dialectical Behavior Therapy (DBT) Skills training, a robust and effective method for improving ER as a tool to change maternal ER in mothers with BPD. By completing multiple assessments of biobehavioral markers of child and mother ER, this study is poised to uncover a potentially modifiable pathway by which these offspring are at risk. Specifically, we propose to conduct a stratified randomized controlled trial of DBT Skills for mothers with BPD who have preschoolers. A total of 300 dyads (initial child age: 36-48 months) will be collected in Eugene, Oregon and Pittsburgh Pennsylvania, with 100 pairs in each of three groups: children who have mothers with BPD who receive DBT Skills, children who have mothers with BPD who receive Family Services as Usual, and children who have non-disordered mothers, matched on income, with this final group only participating in assessments (as to quantify normative ER growth). All children will be assessed 4 times every 4-months, with the first assessment occurring prior to treatment assignment. We will use a biobehavioral laboratory battery to measure child ER, assessing emotion understanding, strategy use, attention regulation and inhibitory control, and parasympathetic regulation. In mothers, we will employ a similar multi-method approach to assess emotion acceptance, strategy use, recognition, and parasympathetic regulation. Finally, during each laboratory assessment, we will observe mother's ability to effectively respond to children in the context of child negative emotions. Growth curve modeling will chart child ER trajectories for all 3 groups so that we can compare child ER growth as a function of: group status (DBT Skills vs. Family Services as Usual vs. income matched, non-disordered controls), changes in maternal ER, and changes in maternal emotion socialization. Findings from this proposal will identify a modifiable pathway by which offspring of mothers with BPD are at risk, determine the extent to which child ER can be restored by treating mothers, and will be the first DBT Skills trial to measure outcomes in offspring.